1. Field of the Invention
The invention relates to an electronic filing system for storing document images into a memory medium and for retrieving document images from the memory medium.
2. Related Background Art
At present, in association with the spread of computers and computer peripheral apparatuses such as scanner, printer, and the like, since it is necessary to effectively use office space and to effectively share and use data, there are many opportunities for analog information on paper, which currently accounts for most office information, to be converted into digital data and stored into a memory device such as a hard disk, magnetooptic memory device, or the like, and then to be used as document data with a computer.
Also, however, document data created by general application software such as a word processor or a DTP (desk top publishing) program on a computer, its amount is increasing more and more in association with the widespread of the personal computer.
Therefore, there are many opportunities such that an electronic filing system or a document management system is introduced for the purpose of managing a large quantity of document data.
In such a situation, how to retrieve target document data from the large quantity of document data which was registered and managed becomes a large problem in document management.
This being the case, a document retrieval method of a conventional electronic filing system shown in FIG. 16 will now be described.
When an instruction for retrieval is issued by the user (step 801), retrieval condition is first set. The setting of the retrieval condition is divided into a setting of text retrieval condition (802) to retrieve from text data in the stored document and a setting of retrieval condition (803) by document attributes such as document name, document creation date, document creator, document keyword, or the like. When the retrieval condition is set, the retrieval is executed by an instruction to execute the retrieval from the user.
In an execution of the text retrieval (804), for the text data in each document managed by an electronic filing system, the text data is sequentially compared with a retrieval character train inputted in step 802 or, if index data of the text data has been formed, the text data is collated with text index data, thereby retrieving the text.
In an execution of the retrieval by document attributes (805), a retrieval is performed by comparing document attribute data managed by the electronic filing system and the retrieval condition set in step 803, thereby judging whether the document attribute data coincides with the retrieval condition or not.
The text which coincides with the retrieval conditions as results of the text retrieval and document attribute retrieval is displayed on a display (806). One retrieval is now finished (807).
However, according to the construction shown in the above conventional technique, the management unit of the document image data digitized by an image input apparatus is nothing but the data of one page unit. In the case where areas of various attributes such as "title", "nature image", "line diagram", etc., are included in one page, the information or data for each of those areas cannot easily be effectively used. Particularly, in the case of a newspaper, magazine, or the like, since many photographs and the like as well as characters are included in one page, using one page as a unit is a data management method which is not sufficiently flexible in terms of the reuse of data and the reconstruction of data.
According to the retrieval method shown in the above conventional technique, for example, in the case where the user cannot remember a character train included in a document to be retrieved, it is not easy to retrieve the target document.
In such a case, even where, for example, an image and a chart are included in the document and the user remembers that fact, the conventional retrieval method provides no means for retrieving by effectively using document construction data such as the image or chart, and it is impossible fully to meet demand for retrieving by various methods for each user.
Further, it will be obviously understood that position information of the document construction data indicating at which position in the page the image or chart is located, is also not used for retrieval.
As will be understood from the situation of the above conventional retrieval method, information which can be used for retrieval includes only the text information in the document or the data added as document attributes later. The construction information of the document indicating at which position in a page the rectangular area of a certain attribute in the document such as an image, chart, or the like is located, is not used for retrieval. A sufficient retrieval method is not afforded to the user.